Serving advertisements based on partial queries

Abstract

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for serving advertisements based on partial queries. In one aspect, a method includes receiving a request for advertisements that specifies a set of query suggestions. The query suggestions are ranked based, at least in part on a probability of each query suggestion being selected by a user that input the partial query. A proper subset of the query suggestions is selected and includes a highest ranked query suggestion based on the ranking and one or more query suggestions of the set of query suggestions that is at least one of a revision of the highest ranked query suggestion or refers to a same topic as the highest ranked query suggestion. One or more advertisements are identified for each query suggestion of the proper subset. Data specifying the identified advertisements are provided.

What is claimed is: 
     
         1 - 20 . (canceled) 
     
     
         21 . A method performed by data processing apparatus, the method comprising:
 receiving, from a search interface presented at a user device, a request for content items, the request specifying a set of query suggestions, the set of query suggestions being identified for a partial query received from a user device;   ranking the query suggestions based, at least in part on a probability of each query suggestion being selected by a user that input the partial query and based on a commercial value score for each query suggestion;   selecting a proper subset of the query suggestions, the proper subset including a highest ranked query suggestion based on the ranking and one or more query suggestions of the set of query suggestions is at least one of a revision of the highest ranked query suggestion or refers to a same topic as the highest ranked query suggestion;   identifying one or more content items for each query suggestion of the proper subset; and   providing an update comprising the identified content items at the user device.   
     
     
         22 . The method of  claim 21 , wherein the commercial value score for each query suggestion is based on a number of content items available for the query suggestion. 
     
     
         23 . The method of  claim 21 , wherein the commercial value score for each query suggestion is based on performance of content items for the query suggestion. 
     
     
         24 . The method of  claim 21 , further comprising, for each partial query, determining a commercial value score for the query suggestions identified for the partial query. 
     
     
         25 . The method of  claim 21 , wherein the commercial value score for each query suggestion is specific to the partial query. 
     
     
         26 . The method of  claim 21 , wherein the commercial value score for one of the query suggestions is further based on at least one of a number of content items available for query suggestions that reference a same topic as the one query suggestion and performance of content items for the query suggestions that reference a same topic as the one query suggestion. 
     
     
         27 . The method of  claim 21 , wherein selecting the proper subset of the query suggestions comprises:
 identifying the highest ranked query suggestion based on the ranking;   determining which query suggestions of the set of query suggestions are related to the highest ranked query suggestion by referring to a same topic as the highest ranked query suggestion or being a revision of the highest ranked query suggestion; and   selecting as the proper subset of query suggestions the highest ranked query suggestion and the query suggestions related to the highest ranked query suggestion.   
     
     
         28 . The method of  claim 27 , wherein determining which query suggestions of the of the set of query suggestions are related to the highest ranked query suggestion comprises:
 determining a similarity measure for each query suggestion of the set of query suggestions, the similarity measure being a measure of the similarity of the query suggestion and the highest ranked query suggestion; and   determining that each query suggestion having at least a threshold similarity measure is related to the highest ranked query suggestion.   
     
     
         29 . The method of  claim 28 , wherein the similarity measure for each query suggestion is based on term overlap between the query suggestion and the highest ranked query suggestion. 
     
     
         30 . The method of  claim 21 , further comprising providing instructions that cause the user device to present the one or more content items in clusters, each cluster being for an individual query suggestion and including content items provided for the individual query suggestion. 
     
     
         31 . The method of  claim 30 , wherein the instructions further cause the user device to present a label for each cluster, the label for a cluster identifying the individual query suggestion corresponding to the cluster. 
     
     
         32 . The method of  claim 21 , further comprising:
 receiving query data defining the partial query, the partial query being query input from the user device and having one or more characters ordered in an input sequence that defines an order in which the one or more characters were input as the query input; and   identifying the set of query suggestions based on the partial query.   
     
     
         33 . A system, comprising:
 a data store for storing content items;   one or more processors configured to interact with the data store, the one or more processors being further configured to perform operations comprising:
 receiving, from a search interface presented at a user device, a request for content items, the request specifying a set of query suggestions, the set of query suggestions being identified for a partial query received from a user device; 
 ranking the query suggestions based, at least in part on a probability of each query suggestion being selected by a user that input the partial query and based on a commercial value score for each query suggestion; 
 selecting a proper subset of the query suggestions, the proper subset including a highest ranked query suggestion based on the ranking and one or more query suggestions of the set of query suggestions is at least one of a revision of the highest ranked query suggestion or refers to a same topic as the highest ranked query suggestion; 
 identifying one or more content items for each query suggestion of the proper subset; and 
 providing an update comprising the identified content items at the user device. 
   
     
     
         34 . The system of  claim 33 , wherein the commercial value score for each query suggestion is based on a number of content items available for the query suggestion. 
     
     
         35 . The system of  claim 33 , wherein the commercial value score for each query suggestion is based on performance of content items for the query suggestion. 
     
     
         36 . The system of  claim 33 , the operations further comprising, for each partial query, determining a commercial value score for the query suggestions identified for the partial query. 
     
     
         37 . The system of  claim 33 , wherein the commercial value score for each query suggestion is specific to the partial query. 
     
     
         38 . The system of  claim 33 , wherein the commercial value score for one of the query suggestions is further based on at least one of a number of content items available for query suggestions that reference a same topic as the one query suggestion and performance of content items for the query suggestions that reference a same topic as the one query suggestion. 
     
     
         39 . A computer storage medium encoded with a computer program, the program comprising instructions that when executed by data processing apparatus cause the data processing apparatus to perform operations comprising:
 receiving, from a search interface presented at a user device, a request for content items, the request specifying a set of query suggestions, the set of query suggestions being identified for a partial query received from a user device;   ranking the query suggestions based, at least in part on a probability of each query suggestion being selected by a user that input the partial query and based on a commercial value score for each query suggestion;   selecting a proper subset of the query suggestions, the proper subset including a highest ranked query suggestion based on the ranking and one or more query suggestions of the set of query suggestions is at least one of a revision of the highest ranked query suggestion or refers to a same topic as the highest ranked query suggestion;   identifying one or more content items for each query suggestion of the proper subset; and   providing an update comprising the identified content items at the user device.   
     
     
         40 . The computer storage medium of  claim 39 , the operations further comprising, for each partial query, determining a commercial value score for the query suggestions identified for the partial query.